Mobile-aided screening system for proliferative diabetic retinopathy

被引:11
作者
Boukadida, Rahma [1 ]
Elloumi, Yaroub [1 ,2 ,3 ]
Akil, Mohamed [2 ]
Bedoui, Mohamed Hedi [1 ]
机构
[1] Univ Monastir, Fac Med, Med Technol & Image Proc Lab, Monastir, Tunisia
[2] Univ Gustave Eiffel, CNRS, ESIEE Paris, Marne La Vallee, France
[3] Univ Sousse, ISITComHammam Sousse, Sousse, Tunisia
关键词
fundus images; mobile health; mobile-aided-screening (MAS) system; neovascularization; random forest (RF); smartphone captured fundus image; RANDOM FOREST; NEOVASCULARIZATION; CLASSIFICATION; SEGMENTATION; IMAGES; RISK;
D O I
10.1002/ima.22547
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
NeoVascularization (NV) occurs in the Proliferative Diabetic Retinopathy (PDR) stage, where the development progress of new vessels presents a high risk for severe vision loss and blindness. Therefore, early NV detection is primordial to preserve patient's vision. Several automated methods have been proposed to detect the NV on retinograph-captured fundus images. However, their employment is constrained by the reduced ophthalmologist-per-person ratio and the expensive equipment for image capturing. This paper presents a novel method for NV detection in smartphone-captured fundus images. The implementation of the method on a smartphone having an optical lens for fundus capturing leads to a Mobiles-Aided-Screening system of PDR (MAS-PDR). The challenge is to ensure accurate and robust detection even with moderate quality of fundus image, on reduced execution time. Within this objective, we identify the major criteria of neovascularized vessels which are tortuosity, width, bifurcation, and density. Our main contribution consists in proposing a sharp feature to reflect each criterion on reduced computational complexity processing. Therefore, the features are provided to a random forest classifier to deduce the PDR stage. A dataset raised from publicly databases is used on a 10-cross validation process where 98.69% accuracy, 97.73% sensitivity, and 99.12% specificity are achieved. To evaluate the robustness, the same experimentation is repeated after applying motion blur filters to the fundus image dataset, where 98.91% accuracy, 96.75% sensitivity, and 100% specificity are deduced. Moreover, NV screening is performed under 3 s when executed in smartphone devices demonstrating the appropriateness of our method to MAS-PDR.
引用
收藏
页码:1638 / 1654
页数:17
相关论文
共 78 条
[1]   Classification of Normal, Ictal and Inter-ictal EEG via Direct Quadrature and Random Forest Tree [J].
Abdulhay, Enas ;
Alafeef, Maha ;
Abdelhay, Arwa ;
Al-Bashir, Areen .
JOURNAL OF MEDICAL AND BIOLOGICAL ENGINEERING, 2017, 37 (06) :843-857
[2]  
Abe S, 2010, ADV PATTERN RECOGNIT, P1, DOI 10.1007/978-1-84996-098-4_1
[3]   A multiscale decomposition approach to detect abnormal vasculature in the optic disc [J].
Agurto, Carla ;
Yu, Honggang ;
Murray, Victor ;
Pattichis, Marios S. ;
Nemeth, Sheila ;
Barriga, Simon ;
Soliz, Peter .
COMPUTERIZED MEDICAL IMAGING AND GRAPHICS, 2015, 43 :137-149
[4]  
Agurto C, 2012, IEEE ENG MED BIO, P4946, DOI 10.1109/EMBC.2012.6347102
[5]   A survey of network anomaly detection techniques [J].
Ahmed, Mohiuddin ;
Mahmood, Abdun Naser ;
Hu, Jiankun .
JOURNAL OF NETWORK AND COMPUTER APPLICATIONS, 2016, 60 :19-31
[6]  
AKIL M, 2018, REAL TIME IMAGE VIDE
[7]   Detection of retinal abnormalities using Smartphone-captured fundus images: A survey [J].
Akil, Mohamed ;
Elloumi, Yaroub .
REAL-TIME IMAGE PROCESSING AND DEEP LEARNING 2019, 2019, 10996
[8]   Detection of neovascularization in retinal images using multivariate m-Mediods based classifier [J].
Akram, M. Usman ;
Khalid, Shehzad ;
Tariq, Anam ;
Javed, M. Younus .
COMPUTERIZED MEDICAL IMAGING AND GRAPHICS, 2013, 37 (5-6) :346-357
[9]  
[Anonymous], 2017, Number of Smartphone Users in the U.S. 2010-2022. Statista DMO
[10]  
Aras R. A., 2016, COMMUN SCI TECHNOL, V1, P33